Check if variable at datetime equal variable at datetime + 24 hours

[Rob Henst]

Good day,

I have two variables (among others): One variable is a binary variable 0 or 1 (sleepwake). The other variable (datetime) is date followed by time (in %tc format). There are 7 days worth of rows, at 15-sec intervals. I want to create a new variable, called "sri", and I want to replace sri for datetime with 0 if sleepwake at datetime != sleepwake at datetime + 24 hours, and I want to replace sri for datetime with 1 if sleepwake at datetime == sleepwake at datetime + 24h. sri for the last day must be "." since there is no sleepwake for datetime + 24 hours to compare with.

datetime sleepwake sri
24feb2017 18:01:30 0 0
24feb2017 18:01:45 0 1
24feb2017 18:02:00 0 1
24feb2017 18:02:15 1 1
24feb2017 18:02:30 1 1
...
25feb2017 18:01:30 1 .
25feb2017 18:01:45 0 .
25feb2017 18:02:00 0 .
25feb2017 18:02:15 1 .
25feb2017 18:02:30 1 .

Please help me find a solution.

Thanks so much.

[Clyde Schechter]

This makes no sense. It is mathematically impossible to have x = x + 24 hours. I suspect you actually have something else in mind, but it isn't clear what that might be. I suggest you post back showing a hand-worked example of what you want along with a step-by-step explanation of how you got it.

In the future, when showing data examples, please use the -dataex- command to do so. If you are running version 15.1 or a fully updated version 14.2, it is already part of your official Stata installation. If not, run -ssc install dataex- to get it. Either way, run -help dataex- to read the simple instructions for using it. -dataex- will save you time; it is easier and quicker than typing out tables. It includes complete information about aspects of the data that are often critical to answering your question but cannot be seen from tabular displays or screenshots. It also makes it possible for those who want to help you to create a faithful representation of your example to try out their code, which in turn makes it more likely that their answer will actually work in your data.

[William Lisowski]

I think I understand that you want to compare a value of sleepwake at a given time with the value recorded precisely 24 hours later. I transformed your sample data into a useful presentation of sample data at some cost in time and tedium. So to offset that time, I offer an incomplete solution that should start you in a useful direction.

* Example generated by -dataex-. To install: ssc install dataex
clear
input double datetime byte sleepwake
1803578490000 0
1803578505000 0
1803578520000 0
1803578535000 1
1803578550000 1
1803664890000 1
1803664905000 0
1803664920000 0
1803664935000 1
1803664950000 1
end
format %tc datetime
tempfile master
save `master'
list, clean
replace datetime = datetime - 1000*60*60*24
rename sleepwake nextday
merge 1:1 datetime using `master'
list, clean

[Rob Henst]

This makes no sense. It is mathematically impossible to have x = x + 24 hours. I suspect you actually have something else in mind, but it isn't clear what that might be. I suggest you post back showing a hand-worked example of what you want along with a step-by-step explanation of how you got it.

In the future, when showing data examples, please use the -dataex- command to do so. If you are running version 15.1 or a fully updated version 14.2, it is already part of your official Stata installation. If not, run -ssc install dataex- to get it. Either way, run -help dataex- to read the simple instructions for using it. -dataex- will save you time; it is easier and quicker than typing out tables. It includes complete information about aspects of the data that are often critical to answering your question but cannot be seen from tabular displays or screenshots. It also makes it possible for those who want to help you to create a faithful representation of your example to try out their code, which in turn makes it more likely that their answer will actually work in your data.

Thanks Clyde, my apologies for not using the right format. I installed dataex (running Stata 12) and will use it next time. With regards to my problem, William Lisowski explains it correctly: I want to compare a value at a specific time, with the value recorded exactly 24h later. I would post the requested example and explanation if William Lisowski hadn't already solved my problem.

Thanks so much again, I really appreciate it.

[Rob Henst]

I think I understand that you want to compare a value of sleepwake at a given time with the value recorded precisely 24 hours later. I transformed your sample data into a useful presentation of sample data at some cost in time and tedium. So to offset that time, I offer an incomplete solution that should start you in a useful direction.

* Example generated by -dataex-. To install: ssc install dataex
clear
input double datetime byte sleepwake
1803578490000 0
1803578505000 0
1803578520000 0
1803578535000 1
1803578550000 1
1803664890000 1
1803664905000 0
1803664920000 0
1803664935000 1
1803664950000 1
end
format %tc datetime
tempfile master
save `master'
list, clean
replace datetime = datetime - 1000*60*60*24
rename sleepwake nextday
merge 1:1 datetime using `master'
list, clean

Thanks so much William. This is exactly right. I apologise for the confusion and for not using the right example data format. Thank you for taking the time and effort to bring the data in a useful format. Offering an incomplete solution for that only makes sense to me.

Just for completeness, should someone else be looking for this solution, I have made the following changes to get what I needed:

tempfile master
save `master'
list, clean
replace datetime = datetime - 1000*60*60*24
rename currentday nextday // I liked your "nextday" varname, so I decided to use "currentday" instead of "sleepwake". 
merge 1:1 datetime using `master' // I really like this approach. 
list, clean

drop status2
drop if currentday==.
drop if nextday==.
drop _merge

gen sri=. 
replace sri=1 if nextday==currentday
replace sri=0 if nextday!=currentday

Thanks so much!!

[William Lisowski]

Glad you were able to take it the rest of the way. Here's a Stata tip inspired by your code.

Logical operators like "==" are no different than arithmetical operators like "+". They take their two arguments and return a value. For logical operators, the value is 0 if the expression formed by the operator and its arguments is false, and 1 otherwise. Since Stata treats 0 as "false" and non-zero as "true" these work as true and false in subsequent expressions using logical operators.

So the last three lines of your code could be replaced by

generate sri = nextday==currentday

and later on, of course, you can use sri in code like

summarize sri

as well as in further logical expressions.